1. Mutation rates are difficult to measure in riboviruses, the largest class of human pathogens. In particular, the question of whether riboviruses share a common genomic mutation rate has resisted resolution. After many efforts, we established a model system using the phage Q growing on the bacterium Escherichia coli. As expected from a ribovirus, mutation rates are very high, about 0.04 per genome replication. The ratios of different kinds of mutations differ from those seen in most other organisms, with a particularly high ratio of transition to transversion base substitutions and with very few indels with a bias towards single-base insertions. We were also able to test the hypothesis that riboviruses replication by a stamping-machine process rather than by exponential replication, with the stamping-machine hypothesis supported. 2. In recent years, several groups have characterized spontaneous mutations using whole-genome DNA sequencing. The methods have potential pitfalls and the results sometimes appear to contradict previous reports using more fully established methods. An example of such a contradiction was described and attributed to the unrealized loss of mutations during the passage of the cultures.